TLC screen for maysin, chlorogenic acid, and other possible

View Sections. ACS2GO © 2018. ← → → ←. loading. To add this web app to the home screen open the browser option menu and tap on Add to hom...
0 downloads 0 Views 425KB Size
J. AMC. Food Chem. 1992, 40, 1211-1213

1211

TLC Screen for Maysin, Chlorogenic Acid, and Other Possible Resistance Factors to the Fall Armyworm and the Corn Earworm in Zea mays Richard C. Gueldner,*J Maurice E. Snook,? Neil W. Widstrom,* and Billy R. Wiseman' Phytochemical Research Unit, R. B. Russell Agricultural Research Center, Agricultural Research Service, U.S.Department of Agriculture, Athens, Georgia 30613, and Plant Resistance/Germplasm Enhancement Research, Georgia Coastal Plains Experiment Station, Agricultural Research Service, U.S.Department of Agriculture, Tifton, Georgia 31793 A thin-layer chromatographic method has been developed for the analysis of maysin, a flavonoid Cglycoside, and its analogues, which have been implicated as resistance factors for the corn earworm (Helicouerpa zea, formerly Heliothis zeal and the fall armyworm (Spodoptera frugiperda). Boric acid present on the TLC plate enhances the fluorescenceof luteolins, including maysin, and increases sensitivity to these compounds by a factor of 10 or more. Another resistance factor, found in smaller amounts, chlorogenicacid, interferes with maysin unless the developingsolvent is made basic with dilute ammonium hydroxide or the TLC plate is treated with boric acid. Leaves, the major food source of the fall armyworm on corn, were not as readily analyzed because of interference due to streaking of the TLC plate.

INTRODUCTION The corn earworm and the fall armyworm are major corn pests in the southeastern United States. Both insects feed in the whorl of young corn plants. As the plant develops, the corn earworm concentrates ita feeding on the silk and the ear. The fall armyworm damages all parta of the plant, but much of the damage arises from leaffeeding. Two metabolites from corn silks that have been shown to be inhibitory to the corn earworm in bioassays are maysin, a flavonoid C-glycoside with a luteolin aglycon (Elliger et al., 1980a), and chlorogenic acid (Elliger et al., 1981). Wiseman et al. (1990)presented evidence that the resistance of centipede grass to the fall armyworm is due to the presence of chlorogenic acid, maysin, and other luteolins. Maysin was originally isolated from Zapalote Chic0 com silks and identified by Elliger et al. (1980a). Snook et al. (1989) developed a high-performance liquid chromatography (HPLC) procedure to quantitate maysin and chlorogenic acid in corn silks. Prior to the development of the quantitative HPLC method for silks, extensive HPLC analyses of leaves of teosinte and many cultivars, inbreds, and populations of corn showed the presence of other luteolii derivatives as well as maysin along with chlorogenic acid and its isomers. Sufficient quantities of the other luteolins have not as yet been isolated for evaluation in bioassays. Genetic studies aimed a t defining the quantitative genetics involved in the biosynthesis of maysin require the analysis of thousands of samples, and for this task HPLC is most suitable. However, screening of a large number of plants for purposes of selection may not require the analytical detail that HPLC provides. Thus, it seemed desirable to develop a thin-layer chromatographic (TLC) method of maysin and chlorogenic acid analysis. While the TLC method would not be as quantitative as the HPLC procedure, it could serve as a rapid screen to identify planta containing genetically amplified levela of maysin and chlorogenic acid. The method could quickly identify promising germplasm and thus reduce or eliminate many of the more Russell Agricultural Research Center.

* Georgia Coastal Plains Experiment Station. f

This

article not subject to U.S. Copyright.

expensive and time-consuming HPLC analyses. We now report our findings regarding TLC analyses of corn leaves and silks for maysin, other luteolins, and chlorogenic acid as a rapid, inexpensive method of screening for insect inhibitory compounds in corn. MATERIALS AND METHODS Plant Materials. Corn leaf and silk samples were obtained in 1988and 1989from two locations: field-grown in Tifton, GA, and greenhouse-grown in Athens, GA. Collection, Treatment, and Storage of Plant Material. Silks (2-5 g/ear) were collected from individual ears and placed in methanol (100mL/silk) in wide-mouthjars, stored at 0 O C , but allowed to warm to room temperature just prior to sampling for HPLC or TLC analysis. For immediate analysis of smaller silk samples without storage, 0.5 g of silks was added to 10 mL of methanol. The mixture was ground and allowed to stand overnight or,alternatively,was ultrasonicatadfor 20 min. Leaves were collected at the 10to 14 leaf stage and were lyophilizedand ground in a Wiley mill before storage at -20 O C in glass or polyethylene bottles. The dried leaf materials were ultrasonicated in the extraction solvent (5050 methanokwater) for 30 min and filtered to produce samples for HPLC. High-Performance Liquid Chromatography (HPLC). Methanol ueed in extractions was Burdick and Jackson (Muskegon, MI) 'distilled-in-glass" grade. Silk extracta,filtered through a nylon 66 membrane fiiter (Snook et al., 1989),were analyzed by reversed-phase HPLC performed with a Hewlett-Packard 1084Bliquid chromatograph,on a 5-pm Altex Ultrasphere ODS column (4.6 mm i.d. X 25 cm). A diode array UV detector monitored samples at 340 nm. We used a linear gradient from methanovwater (2080 v/v, solvent A) to MeOH (solvent B) in 40min with a flow rate of 1mL/min and a 12-min equilibration time. Both solventscontained0.1 % HsPO4. Leaf extracts were analyzed by reversed-phase HPLC using a Hewlett-Packard 1090Mliquid chromatographequippedwith a 1040Adiode array detector and ChemStation data collection system. A Nucleosil 5-pM ODs, 4.6 mm i.d. X 25 cm, column was used for the analysis. The dried leaf materials were ultrasonicated in the extraction solvent (50:50 methanol/water) for 30 min and fiitered through fiiter paper packed tightly into a disposable pipet for HPLC. A gradientstarting at 20% methanol and ending at 100% methanol in 40 min was used, with a flow rate of 0.8 mL/min optimized for resolution according to the procedure of Meyer (1986). Both methanol and water contained 0.1 % HsPO4. Thin-Layer Chromatography (TLC). TLC plates were Whatman silica gel, 260-pm coating on an aluminum backing

Publlshad 1992 by the Amerlcan Chemical Soclety

1212 J. Agric. Food Chem., Vol. 40, No. 7, 1992

with fluorescent indicator at 254 nm (catalog no. 4420 222) or silica gel on glass plates (Merck, art. 5629). Samples were applied directly, without filtering, to the thin-layer plates. For TLC, 1 pL of each sample or standard solutionwas applied to the plate, which was then developed to a heightof 7.5 cm for the aluminumbacked plates or to 3.5 cm for the glass plates with ethyl acetate/ 0.0257'% ammonium hydroxide/methanol50:5:13(solvent M)or ethyl acetate/methyl ethyl ketone/formic acid (88%)/water 5:3:1:1 (solvent N Stahl and Schorn, 1965). The separated compounds on the aluminum-backed plates were visualized by spraying with sodium molybdate reagent for o-dihydroxy compounds (Amow, 1937) or with acidified 1% ferric chloride in water, or by long-wavelengthUV (360 nm) for glass TLC plates sprayed with 10% boric acid in methanol after development or sprayed with 1% boric acid before development. Also, under short-wavelengthUV light the maysin spots showed as dark spots against a yellow-green fluorescent background. Photography for Permanent Record of TLC Plates. Permanent photographic records were made on 35-mm slides of glass TLC plates sprayed with 1% boric acid in methanol and dried to impregnatethe plate with boric acid before development or sprayed with 10% boric after development in solvent N. The 35-mm slides were made wing Ektachrome Daylight ASA 200 film and a 2B UV gelatin filter. Exposures were best atfl.4, '/la s with the glass TLC plate resting directly on top of the UV lamp (long wavelength, 360 nm). RESULTS AND DISCUSSION

Elliger et al. (1980b) examined the structural factors governing toxic effects of flavonoids toward the corn earworm and found that flavonoids, and 0-and C-glycosides of flavonoids, with adjacent hydroxyls or, as in the case of Zea mays,0(3',4')-dihydroxy substitution (luteolin) on the phenyl (B) ring were more potent in bioassays of growth inhibition than were the monohydroxy (apigenin, Le., 4') derivatives. Both maysin and chlorogenic acid fulfill the o-dihydroxy structural criterion, and both are inhibitors of the corn earworm (Elliger et al., 1981; Isman and Duffy, 1982) and the fall armyworm (Wiseman et al., 1990) in bioassays. We presume that the other luteolin derivatives (probably C-glycosides), as yet unidentified and not availablein sufficientquantity for bioassay,will be as active as maysin. Prior to the elucidation of the structure of maysin by Elliger (1980a), Levings and Stuber (1971) reported the occurrence of nonbrowning and browning of silks of corn when the silks were wounded. The browning of the silks was attributed to the presence of substrate luteolinswhich were postulated to enzymatically oxidize to o-quinone compounds that reacted with protein to form the brown pigment. In the nonbrowning mutants the substrate luteolins, not the oxidative enzyme, were lacking. The recessive allele governing synthesis of the luteolins in the silks was independent of luteolin synthesis in the leaves. This finding is consistent with the generally accepted view that flavonoid glycosides are synthesized on the endoplasmic reticulum and then compartmentalized in cellvacuoles and are not transported from leaves to other organs. Therefore, it is expected that leaf and silk are independent in luteolin synthesis, and analyses of both silks and leaves are desirable for selection of plants resistant to both the corn earworm and the fall armyworm. The major flavonoid in most of the corn silks and leaves examined was maysin as indicated by HPLC retention times and UV spectra taken "on-the-fly". Maysin accounted for about 6 0 4 0 % of the detector signal at 340 nm over the range of silks examined. Leaves, however, contained a greater diversity of compounds (Gueldner et al., 19911, representing C-glycosides of luteolins and apigenins and chlorogenic acid and its isomers. Generally, the maysin content was lower in leaves than in silks.

Gueklner et al.

Table I. Color and Adjusted Rf(X100) of Apigenin6 and Luteolins with Boric Acid (BA) Rf compound noBA withBA ARp colorb EM1 (luteolin) 33 27 -6 lemonyellow rutin 44 35 -9 rust EM0 (luteolin) 55 39 -16 light yellow teo apigenin 66 53 -13 lightbrown post apigenin 63 53 -10 blue-green maysin (luteolin) 55 52 -3 yellow premayein (luteolin) 65 55 -10 lemon yellow chlorogenic acid 53 21 -32 blue-violet ARj = no Ba Rf minus BA Rf. * At 360 nm, with boric acid sprayed and dried on plate prior to development.

To separate the diversecomponents of the more complex samples by TLC, 20 solvent combinations were tried with the solventstetrahydrofuran,acetonitrile,methyl and ethyl alcohols, and methyl, ethyl, and isopropyl acetates. Resolution of the flavonoid compoundswas accomplished only with the acetates present as a major component of the solvent system. Separation of maysin, premaysin, and related flavonoids of teosinte leaves was achieved. Chlorogenic acid was not resolved well from maysin in ethyl acetate/methyl ethyl ketone/formic acid/water (5:3: 1:1, solvent N). However, interference with maysin by chlorogenic acid, present in both silks and leaves, was eliminated by substituting 0.025 % ammonium hydroxide for formic acid. The chlorogenic acid spot now ran much closer to the origin (Rf= 0.151, while the maysin spot was relatively unaffected (Rf= 0.52). However, with actual samples streaking obscured the separation of maysin and adjacent spots more than with the formic acid system. Similarly, TLC plates sprayed with 1%boric acid and dried prior to development also separated chlorogenicacid from the luteolin compounds, although streaking reduced resolution of the spots as with the ammonium hydroxide system. Boric acid, however, had the advantage of increasing the fluorescence of the spots, especially enhancing the detectability of the biologically important luteolin derivatives, producing an intense yellow color at 360 nm (see Table I). The maysin spot typically had an Rfof 0.52 and a yellow fluorescentglow under long-wavelength(360 nm) UV light. Related luteolins had a nearly identical color and could readily be distinguished from apigenin8 and chlorogenic acid. Fading of the fluorescence occurred within 15 min, especially if the spots were weak; however, the best quantitation estimates appeared to be with weaker spots. It may be desirable to dilute concentrated samples and to photograph the developed plate as quickly as possible for permanent record. Detectability of maysin with boric acid on the TLC plate prior to solvent developmentwae possible for concentrations as low as 10ng/rL for maysin standard solutions. However, with actual samples this level may not be detectable due to interfering substances, but the level of detectability is estimated to be in the range 10-20 ng/rL. Boric acid also did not interfere with detection of spots under short-wavelength UV (254 nm). Color development (Table I) was achieved by treatment ' ferric with Amow's reagent (Amow, 1937)and acidified 1% chloride in water. Amow's reagent produced a bright yellow color for the o-dihydroxycompounde;ferric chloride gave a gray-brown color but was not selective for the odihydroxy compounds. Boric acid caused a strong fluorescence witht he luteolins when it was sprayed onto the TLC plate before solvent development. The fluorescent effect was similar but not as uniform when the plate was sprayed with boric acid after solvent development.

J. Agrlc. FoodChem., Vol. 40, No. 7, 1992

TLC Screen for Maysln

and 254 nm. The limit of detectability here, without boric acid, was 10times greater, and the estimates were high by 2-3-fold at stronger maysin concentrations. Chlorogenic acid was too low in these silk samples to be detected by TLC in the samples of both Tables IIA and 111. Chlorogenic acid is usually at low levels in the silk and thus is not often a significant factor in conferring resistance to feeding by insect larvae. Levels of chlorogenic acid tend to be more prominent in leaves than in silks but not as prominent as maysin and its relatives. This TLC method provides a fast, inexpensive screen of large numbers of corn silks so that promising crosses containing high levels of maysin or other luteolin derivatives may be identified quickly. The method has sufficient sensitivity to differentiate levels of maysin necessary to impart resistance to the corn earworm, estimated to be 0.2 % or greater of the fresh weight of silk (Snook et al., 1989). For leaves, however, confounding factors may interfere with maysin, and estimates of rank by TLC do not always correlate well with the HPLC method.

Table 11. Visual Ranking of Order of Maysin Amount in Silk and Leaf Samples Analyzed by TLC. Compared with HPLC Amounts and Ranking sample 1127A

B C D E F G H I J

rank by amt of mayain HPLC TLC spotted,”ng A. SilkMayain 1 1 106.2 2 2 73.6 3 3 48.2 4 4 38.5 5 5 31.8 6 6 31.4 7 7 23.4 8 8 19.6 9 9 0.5 10 lad 0.0

B. LeafMayain K L M N 0 P

B

1 2 3 4 5

6 7 8 9

2 1 4 8 3 9 5

990 930 73 54

45 39 33 28 20

%

maysine 0.219 0.320 0.201 0.135 0.148 0.131 0.066 0.056 0.003 0.000 2.05 2.21 0.23 0.16 0.15 0.09 0.09 0.08 0.06

ACKNOWLEDGMENT

7 R S 6 a Fluorescence induced by boric acid at 360 nm. Amount of maysin calculated from HPLC analyses. Percent mayein,by freshweight, as determined by HPLC. d No maysin detected.

We are grateful to Susan J. Wilson for her excellent technical support.

Table 111. Visual Quantitation of Maysin in Corn Silks by maysin, pg chlorogenic acid, silk maysin, estde calcdb % fresh wtb % freshwtb sample 1.3 0.7000 0.026 1.040 256-2 1.5 0.4800 0.019 1.010 256-1 0.8 0.4100 0.030 0.740 256-8 0.5 0.2900 0.029 0.610 256-5 0.3 0.2800 0.027 0.570 232-1 0.4 0.2800 0.035 0.420 256-3 0.3 0.1700 0.012 0.330 256-12 0.2 0.2100 0.064 0.260 238-8 0.0 0.0120 0.011 0.011 816-9 0.0 0.0028 0.001 0.005 810-3 0 The silica gel plate with fluoreecent background was developed with solvent M. * By HPLC. e By TLC under short wavelength UV (254 nm).

The application of the TLC method without boric acid and with visualization under short wavelength to corn silks is summarized in Table 111. Estimates of concentration are too high when intense spots are evaluated, while lower intensity spots gave estimates much closer to the values determined by HPLC. With leaf samples (Table IIB) it was possible to rank in order the amounts spotted as calculated from the HPLC analyses. Table I1 compares the highest to lowest ranking by HPLC to the TLC ranking. The rankings matched perfectly, but in the lowest two samples (I and J) no maysin could be detected. Thus, detectability in actual silk samples is as low as 20 ng/pL. In Table I11 estimates of silk amounts of maysin were made bv comparison with known standard concentrations of maysin using the fluorescent background quenching

1218

*

LITERATURE CITED Amow, L. E. Colorimetric determination of the components of 3,4dihydroxyphenyWietyroeine mixtures. J.Biol. Chem. 1937,118,531-537. Elliger, C. A,; Chan, B. G.; Waiss, A. C., Jr.; Lundin, R. E.; Haddon, W. P. C-glycosylflavonesfrom Zeamaye that inhibit ineeCt development. Phytochemistry 1980a, 19, 293-297. Elliger, C. A.; Chan, B. G.; Waiss, A. C., Jr. Flavonoids as larval growth inhibitors. Naturwissenschaften 1980b, 67,358-60. Elliger, C. A.; Wong, Y.; Chan, B. C.; Waiss,A. C., Jr. Growth inhibitors in tomato (Lycopersicon)to tomato fruitworm (Heliothis zea). J. Chem. Ecol. 1981, ?, 753-758. Gueldner, R. C.; Snook, M. E.; Wiseman, B. R.; Widetrom, N. W.; Himmelsbach, D. S.; Costello, C. E. Maysin in corn, teosinte, and centipede grass. In Naturally OccurringPest Bioregulators; Hedin, P. A., Ed.; ACS Symposium Series 449; American Chemical Society: Washington, DC, 1991. Isman, M. B.; Duffey, S. S. Phenolic compounds in foliage of commercial tomato cultivars as growth inhibitors to the fruitworm, Heliothis zea. J. Am. SOC.Hortic. Sci. 1982, 107 (l), 167-170. Levings, C. S.; Stuber, C. W. A maize gene controlling silk browning in response to wounding. Genetics 1971,69,491498. Snook, M. E.; Widstrom, N. W.; Gueldner, R.C. A reversed phase, high performance liquid chromatographic procedure for the determination of maysin in corn silks. J. Chromatogr. 1989, 477,439-447. Stahl, E.; Schom, P. J. In Thin Layer Chromatography: A Laboratory Handbook; Stahl, E., Ed.; Academic: New York, 1965; p 376. Wiseman, B. R.; Gueldner, R. C.; Lynch, R. E.; Severson, R.F. Biochemicalactivity of centipedegrw against fall armyworm larvae. J. Chem. Ecol. 1990,16 (9), 2677-2690. Received for review October 30, 1991. Revised manuscript received March 30,1992. Accepted April 20,1992.